Hi,
Before I start, I realize the best solution may be to replace my RX 470, however I'm currently not in a financial position to do so..
I bought my setup 2nd-hand and it was working well for 2 months, then I started to have a couple of random power shutdowns. I checked the psu and it was an old 265W DELL unit, and given that a 400W supply is recommended for this GPU I assumed this was the problem. Running a power test using OCCT caused instant power shutdown, whereas all other tests ran without problem.
I purchased and installed a new 500W PSU and the problem persisted (instant shutdown with OCCT power test), it was at this point I realized one of the GPU fans was not spinning and the shutdown was due to a sudden spike in GPU temperature. I manually played with the fan speeds using the AMD software and realized one fan would not spin, and the other would only spin when set to 100% speed - requiring me to set 100% fan speed at 60 C. This resolved the issue somewhat, but was obviously not ideal so I removed the GPU to inspect the fans. I disconnected the fans from the GPU and attached a 9V battery to test them, both immediately spun at high speed - suggesting they are fine (?). I cleaned the GPU and reinstalled it..
On powering up, the screen resolution had dropped from 4K down to 600p (?) and could only be increased to 1024p (?). I can't remember exactly and I'm at work at the moment and can't check, but they were non-standard resolution values. Running the AMD software resulted in an error stating that the required driver was not installed or functioning properly. On ispection of device manager there was an error for the PCI port stating no driver was found, and the GPU had the "Windows has stopped this device because it has reported problems. (Code 43)" (neither were there previously).
I attempted a system restore (to before physically removing the GPU) which had no effect on the driver issues. I then used a driver scanning utility, which recognized the missing PCI port driver and installed it, but the GPU error persists. I've scanned the setup using AMD software and re-installed the current driver, however it has no effect.
Any ideas what is going on? How could I have had this effect by just physically removing and reinstalling the GPU?.. The system was running stably before I chose to remove the GPU for inspection.
